Abstract
Parkinson's disease (PD) is a progressive neurodegenerative disorder marked by the loss of dopaminergic neurons and widespread transcriptomic dysregulation across disease stages. Patients commonly exhibit motor symptoms such as tremors, rigidity, and bradykinesia, alongside non-motor symptoms including depression and cognitive decline. While previous research has largely focused on protein-coding genes, growing attention is being directed toward the regulatory roles of non-coding RNAs in PD pathogenesis-particularly the interplay between circular RNAs (circRNAs) and microRNAs (miRNAs). Emerging evidence indicates that circRNAs can act as competing endogenous RNAs (ceRNAs), modulating gene expression by sequestering miRNAs and thereby mitigating miRNA-mediated repression of target mRNAs. In this study, we performed a dynamic transcriptomic analysis across four PD stages using RNA-seq data to identify differentially expressed circRNA-miRNA-mRNA networks. We constructed stage-specific ceRNA networks by selecting positively co-regulated circRNAs and linear transcripts that were co-expressed exclusively within the same disease stage. Among the upregulated circRNAs with predicted ceRNA activity, circPRDM2 and circHSH2D were identified as uniquely expressed in PD patients. Additionally, we assessed the coding potential of the predicted target genes to further elucidate the regulatory impact of circRNAs on mRNA expression. Our findings provide new insights into the post-transcriptional regulatory mechanisms involved in PD and highlight candidate stage-specific ceRNA axes that may serve as potential biomarkers or therapeutic targets.